JPH0584065B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method of manufacturing a solid-state image sensor.

Conventional solid-state image sensors are broadly classified into line sensors and area sensors, with line sensors being used for reading facsimiles and the like, and area sensors being used for video cameras.

With the recent development of information processing equipment, there is a growing demand for inexpensive and high-performance devices and equipment. In particular, this demand is increasing as devices become more widespread from office use to personal and home use.

For example, facsimile machines for home use that cost less than 200,000 yen are being introduced to the market. In facsimile, the system consists of a readout section, a recording section, and a communication system, but the cost of the recording section has become considerably lower due to the development of thermal heads, etc., and the communication system has become considerably cheaper due to the development of LSI. Although we are nearing completion, the read-out section has a complicated optical system and the sensor itself is expensive, so the overall cost will be high.

Therefore, there is a need for a technology for manufacturing this lead-out section at low cost and with high performance. If it becomes possible to reduce the cost of this part, facsimile,
By organically combining it with copy machines and printers, more advanced equipment can be realized as a versatile machine with intelligent functions.

In order to reduce the cost and improve the performance of this readout section, an image sensor with a simple optical system is required. For this reason, in recent years, close-contact sensors have been proposed in which the object to be read and the image sensor are brought into close contact with each other.

However, in reality, they have had drawbacks such as insufficient characteristics, poor reliability, and external processing that is too complex to be cost-effective.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a solid-state image sensor that has high performance, is sufficiently reliable, and can be manufactured at low cost. A further object of the present invention is to provide an image sensor manufacturing method that improves the characteristics of a thin film transistor attached to a photoreceptor at the same time as the process of forming the photoreceptor.

The present invention provides a method for manufacturing an image sensor in which a thin film photoreceptor and a thin film transistor electrically connected to the thin film photoreceptor and selecting the thin film photoreceptor are formed on a substrate having at least an insulated surface. forming a polycrystalline silicon film on the substrate, which will become the channel portion of the thin film transistor; and after this step, forming an amorphous silicon film, which will become the thin film photoreceptor, on the substrate in a plasma atmosphere containing H ₂ . It is characterized by having a step of forming by CVD method.

FIG. 1 is a block diagram of a line sensor used in the present invention.

N bits of elements 8 are arranged in a line, and one element consists of a scan circuit 1, a switching circuit 2, and a photosensitive cell section 3.

The scan circuit 1 is basically a shift register, and is input to the gate 5 of the switching transistor 4 of the switching circuit 2.
Control ON-OFF.

The basic operation is to read out the discharge amount of the charges stored in the photosensitive cell section 3 to the output line Vo by turning on the switching transistor 4 in accordance with the amount of light irradiated.

The N-bit cells are sequentially read out by the scan circuit and appear on the output line Vo as serial data for each cell. As a result, the amount of light irradiated to each cell is converted into an amount of electricity in proportion to the amount of light.

FIG. 2 is a diagram showing a specific circuit of FIG. 1,
1 is a shift register, 2 is a transistor which is a switching circuit, and 3 is a photosensitive cell.

10 is a clock line, 11 is a data input line, 12 is a flip-flop, 13 is a thin film transistor, 14 is a photoelectric conversion element, and 15 is a capacitor.

At this time, the feature of the embodiment of the present invention is that all elements including transistors are formed of thin films.

Figure 3 shows the operating waveforms of each part of this circuit. When each output Q ₁ to Q _o of the shift transistor array is sequentially output, the charging current increases as the switching transistors are sequentially selected. will appear on the output line. Since this peak value corresponds to the amount of light in each cell, by passing the signal through a low-pass filter or a peak hold circuit, a signal level proportional to the amount of light can be obtained.

Circuits such as switching, transistors, and shift registers used in the present invention use thin film transistors (polysilicon-TFT) made of polycrystalline silicon, and amorphous silicon is used for the photoreceptor film. The advantages of this method are (1) Naturally, light is incident on the switching transistors and peripheral drive circuits, but when light is incident, a photocurrent is generated when the transistor is in the OFF state, causing leakage current. , the transistor will no longer turn off completely, resulting in it stopping or malfunctioning. However, polysilicon-TFT absorbs very little light.
Does not cause malfunction even if light enters it.

(2) On the other hand, for photoconductive films, the higher the light absorption, the better;
Amorphous silicon film is optimal.

(3) Amorphous silicon deposition method
When plasma CVD is performed in an H ₂ atmosphere, H atoms cancel the level at the polysilicon-TFT interface, and the transistor characteristics of the polysilicon-TFT are dramatically improved. In other words, as is well known, amorphous silicon is produced by plasma CVD in a SiH ₄ gas atmosphere.
Hydrogen-containing amorphous silicon is formed, but in such a state the H ₂ of polysilicon-TFT
Cannot be processed. Therefore, by further adding an H ₂ atmosphere, it is possible to perform H ₂ treatment on the TFT while depositing amorphous silicon. Although the polysilicon of the TFT channel contains many dangling bonds, mobility can be improved by filling these dangling bonds with H atoms through H ₂ treatment. Therefore, by simply depositing an amorphous silicon film on a polysilicon-TFT,
This automatically improves the performance of polysilicon TFTs.

(4) Since amorphous silicon is formed in an H ₂ atmosphere, more H elements are introduced into the down bonds present in amorphous silicon, making it easier for photocurrent to flow and improving photosensitivity.

FIG. 4 shows a concrete implementation example in which the thin film transistor is formed of polysilicon and the photoconductive film is formed of amorphous silicon according to the present invention, and A and B show cross sections AB.

Substrate 3 made of material such as glass or ceramics
By depositing and patterning a polycrystalline silicon thin film on 1, a source 34 and a channel 3 are formed.
3. Form a drain 32 region.

Thereafter, a gate film 35 for gate insulation is formed by thermal oxidation or CVD, and a gate electrode material such as polycrystalline silicon is deposited and patterned to form a gate 36. Then, the source and drain electrodes 32 and 33 are formed using the ion implantation method.
Create a P-type or N-type region.

After that, an interlayer insulating film, for example, a silicon oxide film 41
are formed by the CVD method to form contact holes 37 and 43.
Holes are opened to form an Al wiring layer that will become an output line and an Al layer 39 that will serve as a lower electrode of the photosensitive layer. Then, a photosensitive layer 40 made of amorphous silicon or the like is deposited over the entire structure by a plasma CVD method in an H ₂ atmosphere, and a transparent electrode layer 42 made of ITO or the like which becomes the upper electrode of the photoreceptor is formed thereon.

Thereafter, if necessary, the amorphous silicon film and ITO film are etched off leaving only the amorphous silicon film and the ITO film on the electrode 39. The photoreceptor layer 40 has a dark current of less than InA when not irradiated with light, and has a dark current of several nA/in when exposed to light.
Set it to lx. The advantage of this method is that both the photoreceptor and the capacitor are formed. When amorphous silicon is used as the photoreceptor layer 40, the dark current is very small and the photocurrent is large, making it suitable for this optical reading.

FIG. 11 shows a typical example of the photosensitive characteristics of this amorphous silicon film, which is characterized in that it can be used at illuminances of 11x (1 lux) or less.

As shown in Figure 4, the photoreceptor layer is placed vertically (film perpendicular).
The advantage of the conductive type is that there is no need to etch off the photoreceptor layer or upper electrode, and the film can simply be deposited.

FIG. 5 shows another example of the method of the present invention.

This uses a horizontal (horizontal film) conductive type photoreceptor layer. A is a CD cross section of B, and will be explained according to the formation process.

A silicon thin film for forming a transistor and a capacitor is formed on a substrate 51 by the CVD method, and after patterning, a gate oxide film 55 is formed. After that, N or P is applied to the lower electrode portion 54 of the charge product capacitor.
A mold layer is formed by ion implantation.

After that, a gate electrode 56 made of crystalline silicon or the like and an upper electrode 57 of the capacitor are formed, and then ion implantation is performed one more time to form an N or P type source region 52, a negative region channel region 53, and a drain region 61. A switching transistor section consisting of the gate electrode 56, a capacitor consisting of the lower electrode 54, the upper electrode 62 and the insulating film 55 are formed.

After that, an interlayer insulating film 58 is deposited, contact holes 60, 61, and 62 are opened, and an Al wiring 63 that becomes an output line and a photoreceptor layer 59 are formed. The photoreceptor layer is a semiconductor material sensitive to light, such as CdS or amorphous silicon, and is arranged in parallel with the capacitor.

As a result, when no light is irradiated, the photoreceptor layer 5
9 has a very high resistance and does not discharge the charge accumulated in the capacitor, but when it is irradiated with light, it discharges the charge in the capacitor, so when the switching transistor is turned on, a charging current is generated. As a result, the amount of light is converted into an amount of electricity.

The feature of the method shown in FIG. 5 is that by using the photoreceptor film as a horizontal conductor, upper and lower electrodes are not required, and that it can be used even if the film has many pinholes.

6 and 7 are other embodiments of the present invention,
In contrast to the structure shown in FIG. 4, a photoconductive film is formed on a thin insulating film.

That is, when forming the amorphous silicon film 40 that becomes a photoconductive film on the Al electrode 39 in the structure shown in FIG. 4, it is possible to prevent Al from diffusing into the amorphous silicon film and destroying the film. In order to suppress the spread of the depletion layer from the side to the amorphous silicon film and improve the breakdown voltage,
After covering with an insulating film such as SiO ₂ or Si ₃ N ₄ , an amorphous silicon film is formed.

As a result, this insulating film is equivalently inserted as a capacitor 66 between the transistor 65 and the photoconductive film's equivalent resistance 68 and capacitance 67, as shown in FIG. This makes the signal easier to handle.

FIG. 7 shows an example of this structure, showing the source and drain layers 71 and 73 of the transistor on the substrate 70, the channel 72, the gate insulating film 74, and the gate electrode 7.
5 is formed as a polysilicon TFT. Thereafter, after forming an interlayer SiO ₂ film 83, contact holes 76 and 77 for the source and drain portions are opened.
After that, Al, AlSi, or Al-Si-Cu is formed as a smooth film without hillocks, and after etching, the lead wiring 78 and the electrode 7 to the bus line Vo are formed.
form 9.

After this, the insulating film 80 is formed by, for example, atmospheric pressure CVD method.
A SiO ₂ film of about 1000 to 3000 Å is deposited, and then an amorphous silicon film 81 and a transparent conductive film, for example an ITO film 82, are formed.

If necessary, as shown in the figure, only the photoconductive portion is etched off leaving the ITO film 82, and then amorphous silicon is etched using the remaining ITO film as a mask.

In addition, as the insulating film 80, it is made by plasma CVD method.
The process becomes simpler if Si ₃ N ₄ , SiN, or SiO ₂ is formed and the amorphous silicon film 81 is deposited continuously by switching the gas.

The scan circuit used in the present invention is required to have a certain degree of high speed. For example, the number of elements
1000 and the read cycle is 1 msec, the scan speed is 1 MHz. For this reason, the scan circuit uses a shift register that can operate at high speed,
A transistor is required to construct it. For this reason, it is preferable to form the scan circuit with a polysilicon TFT that can increase mobility.

Figure 12 shows an N-
This is an example of a scan circuit using MOS (P channel may also be used).

The cell 116 includes a transfer transistor 111, a write transistor 113, a transistor 112 serving as a bootstrap capacitor, a pull-down transistor ₁₁₅ , and a charge erasing transistor ₁₁₄ . works like this.

The T output or the S output is used as the gate input of the switching transistor of the photosensitive cell. In this method, the bootstrap circuit can be configured with TFT, so
Compared to a bulk MOS transistor, since the substrate is an insulator, the variable ratio of the bootstrap capacitance can be increased, and as a result, the voltage to be stepped up is large, and the rise speed of the S output is improved.

Furthermore, since the transistor 113 does not have a back channel effect unlike a bulk MOS transistor, there is no rise in threshold voltage as the voltage rises, and there is an advantage that the S output can be started up quickly.

The TFT used here can be realized using either a P channel or an N channel transistor shown in FIG.

FIG. 8 is an example of a scan circuit having a C-MOS configuration, and shows one element.

P-channel thin film transistor (P-TFT) 9
0 to 93 and N channel thin film transistor (N-
TFT)94-97.

Figure 9 shows an example of the structure of this CMOS-TFT.
After forming a first silicon thin film 101 on a substrate 100, a gate oxide film 102 is formed, and then a gate electrode 103 is formed.

Thereafter, boron ions are implanted into the P-channel transistor 104, and phosphorus or arsenic ions are implanted into the N-channel transistor 105, thereby forming the respective transistors.

In this way, a major feature of TFTs, compared to conventional single-crystal wafer-based image sensors, is that a monochannel device (N-NOS or P-MOS) can be converted to CMOS by simply adding one ion implantation process. . One reason for this is that whether the channel region is P type or N type, an intrinsic region containing no impurities is commonly used.

The transistor (TFT) used in the present invention is required to be fast both in a scan circuit and as a switching transistor, that is, it is necessary to improve the characteristics of the transistor.

As an example of the process for forming the transistor section used in the present invention, good transistor characteristics can be obtained by using a thermal oxide film as the gate insulating film.

Depressurize the impurity-free silicon thin film that makes up the first layer channel, source, and drain.
Approximately at a deposition temperature of 570℃ by CVD method
After 2000-5000Å formation and patterning, 1100℃
Thermal oxidation in _O2 atmosphere at ~1150℃ to approximately 1500Å
At the same time as forming a good gate insulating film, grains of the first silicon thin film are grown to form a good polycrystal.

Thereafter, a gate electrode of polycrystalline silicon doped with N ⁺ is formed, and then P ions are implanted at a doping amount of 1×10 ¹⁵ /cm ² using the gate electrode as a mask, leaving only the channel as an intrinsic region.

After this, when H ₂ plasma treatment is performed, the characteristics are further improved. When amorphous silicon is used as the photoreceptor film in the systems shown in FIGS. 4, 5, and 7, when hydrogen-based plasma CVD is used, the TFT is automatically subjected to H ₂ plasma treatment at the same time.

It is also possible to separately perform the method shown in FIG. Figure 10 shows an example of the characteristics of an N-TFT obtained through such a process, and the channel carrier mobility is approximately 80 cm ² /V sec, which is a good characteristic that is approximately 1/5 that of single crystal silicon. It is.

Furthermore, even under 1000 lx, light leakage hardly increases, and stable operation is possible even under light irradiation.
A scan circuit constructed using this transistor operates at approximately 2 to 5 MHz, providing sufficient high speed. Furthermore, the switching speed of the switching transistor is 100 nsec or less.

According to embodiments of the present invention, there are the following advantages.

(1) Since it is constructed on an insulating substrate through a simple process, there are no size restrictions like single crystal silicon, making it possible to create a close-contact sensor with a length of 10 cm to 30 cm, resulting in lower costs.

(2) By incorporating a scanning circuit and a switching circuit, only 10 external wiring lines are required, significantly reducing implementation costs. In addition, since the output line is placed on an insulator, stray capacitance is extremely small, the amplitude of the output signal can be maintained up to the power supply voltage used, the S/N ratio is greatly improved, and there is no need for a complicated subsequent amplifier. Sufficient signal level is ensured and the printed image is clear.

(3) Adoption of polycrystalline silicon TFTs as transistors improves switching speed and greatly improves reliability and stability. or
Since it is easy to convert into CMOS, good values for operating speed and power consumption can be obtained when applied to scan circuits. Furthermore, the process is simple and costs can be easily reduced.

(4) Since the photoreceptor layer is made thinner, variations in photoreceptivity due to lifetime distribution, as in single-crystal silicon, are suppressed, and the sensitivity distribution in the sensor's line direction is significantly reduced.

As described above, according to the present invention, there is a step of forming a polycrystalline silicon film, which will become the channel portion of the thin film transistor, on the substrate, and after this step, an amorphous silicon film, which will become the thin film photoreceptor, is formed on the substrate. In addition, the process of forming the amorphous silicon film by a plasma CVD method in an H ₂ -containing atmosphere was performed, so that the polycrystalline silicon film in the channel portion of the thin film transistor that had been previously formed was treated with H ₂ . Further, the hydrogenation of amorphous silicon progresses, and it is possible to simultaneously form an amorphous silicon film and improve its characteristics and improve the characteristics of a thin film transistor without increasing the number of steps.

As described above, the present invention can realize a high-performance, highly reliable, and low-cost solid-state image sensor, and its effects are significant.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a solid-state image sensor used in the present invention. FIG. 2 is a specific circuit diagram thereof. FIG. 3 is a diagram showing operating waveforms of the circuit diagram of FIG. 2. FIGS. 4A and 4B are diagrams showing an embodiment of the present invention. FIGS. 5A and 5B are diagrams showing embodiments of the present invention. FIG. 6 is a circuit diagram of FIG. 7, which is a diagram showing an embodiment of the present invention. FIG. 7 is a diagram showing a specific structural example of the embodiment shown in FIG. 6. 8th
12 are diagrams showing an example of a scan circuit. FIG. 9 is a diagram showing an example of the structure of CMOSTFT. FIG. 13 is a diagram showing the operating waveforms of FIG. 12. FIG. 10 is a diagram showing an example of the characteristics of the N-TFT used in the present invention. FIG. 11 is a diagram showing the optical characteristics of the photoreceptor layer.

Claims (1)

[Scope of Claims] 1. A method for manufacturing an image sensor, in which a thin film photoreceptor and a thin film transistor electrically connected to the thin film photoreceptor and selecting the thin film photoreceptor are formed on a substrate whose surface is insulated at least. forming a polycrystalline silicon film on the substrate, which will become the channel portion of the thin film transistor; and after this step, forming an amorphous silicon film, which will become the thin film photoreceptor, on the substrate in an H ₂ -containing atmosphere. 1. A method for manufacturing an image sensor, comprising: a step of forming the image sensor by a plasma CVD method.